* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Revision of Electromagnetic Theory Lecture 2
History of electromagnetic theory wikipedia , lookup
Speed of gravity wikipedia , lookup
Nordström's theory of gravitation wikipedia , lookup
Coherence (physics) wikipedia , lookup
Equations of motion wikipedia , lookup
Field (physics) wikipedia , lookup
Fundamental interaction wikipedia , lookup
Photon polarization wikipedia , lookup
Gravitational wave wikipedia , lookup
Electromagnetic mass wikipedia , lookup
Lorentz force wikipedia , lookup
Diffraction wikipedia , lookup
First observation of gravitational waves wikipedia , lookup
Maxwell's equations wikipedia , lookup
Introduction to gauge theory wikipedia , lookup
Aharonov–Bohm effect wikipedia , lookup
Time in physics wikipedia , lookup
Theoretical and experimental justification for the Schrödinger equation wikipedia , lookup
Revision of Electromagnetic Theory Lecture 2 Waves in Free Space Waves in Materials Waves in Bounded Regions Generation of Electromagnetic Waves Andy Wolski University of Liverpool, and the Cockcroft Institute Maxwell’s Equations In the previous lecture, we discussed the physical interpretation of Maxwell’s equations: ~ = ρ, ∇·D (1) ~ = 0, ∇·B (2) ~ ~ = −∂B , ∇×E (3) ∂t ~ ~ = J~ + ∂ D , ∇×H (4) ∂t and looked at some applications in accelerators, involving static (time independent) fields. We also saw how the fields could be expressed in terms of the potentials: ~ = ∇ × A, ~ B (5) ~ ~ = −∇ϕ − ∂ A . E ∂t Electromagnetism 1 (6) Lecture 2: Electromagnetic Waves Electromagnetism in Accelerators In this lecture, we shall: • show that Maxwell’s equations lead to wave equations for the electric and magnetic fields, and explore the solutions to these equations in free space; • discuss the behaviour of electromagnetic waves in conducting materials; • derive the solutions to the wave equations in regions bounded by conductors; • consider some applications in accelerators, in particular in waveguides and rf cavities; • discuss (briefly) the generation of electromagnetic waves. 2 Electromagnetism Lecture 2: Electromagnetic Waves Electromagnetic Waves in Free Space In free space, Maxwell’s equations (3) and (4) take the form: ~ ∂B ~ ∇×E = − , (7) ∂t ~ ~ = µ0ε0 ∂ E , ∇×B (8) ∂t where µ0 and ε0 are respectively the permeability and permittivity of free space (fundamental physical constants). If we take the curl of (7), then, using a vector identity, we find: ~ ~ ≡∇ ∇·E ~ − ∇2 E ~ = − ∂ ∇ × B. ∇×∇×E ∂t (9) ~ = 0; then, substituting for ∇ × B ~ In free space, we have ∇ · E from (8) we obtain the wave equation: ~ − µ0ε0 ∇2 E Electromagnetism 3 ~ ∂ 2E = 0. ∂t2 (10) Lecture 2: Electromagnetic Waves Electromagnetic Waves in Free Space A solution to the wave equation (10) can be written: i ~k·~ r−ωt ~ r, t) = E ~ 0e E(~ , (11) ~ 0 and ~k are constant vectors, ω is a constant, and where E ~ r = (x, y, z) is the position vector. Substituting this solution into the wave equation yields the relationship between ω and ~k: ω 1 = √ = c. ~ µ ε k 0 0 (12) Equation (12) gives a relationship between the wavelength ~ λ = 2π/ k and frequency f = ω/2π of the wave; such an equation is known as a dispersion relation. Electromagnetism 4 Lecture 2: Electromagnetic Waves Electromagnetic Waves in Free Space Since we obtained the wave equation by taking additional derivatives of Maxwell’s equations, we should check that the solution (11) satisfies the original equations. In this case, we find that Maxwell’s equation (1) (in free space): ~ = 0, ∇·E (13) ~k · E ~ 0 = 0. (14) ~ 0 and the wave gives a relation between the amplitude vector E vector ~k: ~ 0 gives the direction of the electric field at all points in Since E space, and ~k gives the direction of propogation of the wave, equation (14) tells us that plane electric waves in free space are transverse waves. Electromagnetism 5 Lecture 2: Electromagnetic Waves Electromagnetic Waves in Free Space We can also derive a wave equation for the magnetic field, starting from Maxwell’s equation (8) (in free space): ~ ~ = µ0ε0 ∂ E . ∇×B ∂t Taking the curl, as before, and using a vector identity gives: ~ − ∇2 B ~ = µ0ε0 ∂ ∇ × E. ~ ~ ≡∇ ∇·B ∇×∇×B ∂t (15) ~ = 0, and substituting for ∇ × E ~ from Maxwell’s Using ∇ · B equation (3) gives: ~ − µ0ε0 ∇2 B ~ ∂ 2B = 0, ∂t2 (16) ~ i k·~ r−ωt (17) which has the solution: ~ r, t) = B ~ 0e B(~ 6 Electromagnetism , Lecture 2: Electromagnetic Waves Electromagnetic Waves in Free Space As before, we have to check that our solution for the magnetic wave equation satisfies Maxwell’s equations. In this case, we ~ = 0 that the wave is once again a transverse find from ∇ · B wave: ~k · B ~ 0 = 0. (18) We also have to verify that our solutions for the electric and magnetic fields simultaneously satisfy Maxwell’s equations. We find: ~ ∂B ~ ~ 0 = ωB ~ 0, ∇×E =− ⇒ ~k × E ∂t ~ = µ0ε0 ∇×B ~ ∂E ~ 0. ~0 = ω E ⇒ ~k × B ∂t c2 (19) (20) These results tell us that the electric and magnetic fields and the wave vector must be mutually perpendicular... Electromagnetism 7 Lecture 2: Electromagnetic Waves Electromagnetic Waves in Free Space 8 Electromagnetism Lecture 2: Electromagnetic Waves Electromagnetic Waves in Free Space We also find the relationship between the amplitudes of the electric and magnetic fields: ~ E0 ω ~ = ~ = c. B 0 k (21) ~0 = B ~ 0/µ0: In terms of the magnetic intensity H where we have used: s ~ E0 µ0 = µ0 c = , ~ ε0 H0 c=√ 1 . µ0ε0 (22) (23) ~ and H ~ in a plane wave The ratio between the amplitudes of E in free space defines the impedance of free space, Z0: Z0 = Electromagnetism s µ0 ≈ 376.730 Ω. ε0 9 (24) Lecture 2: Electromagnetic Waves Electromagnetic Waves in Dielectrics Electromagnetic waves in dielectrics generally behave in a similar way to electromagnetic waves in free space, except that the phase velocity vφ is modified by the relative permeability and relative permittivity: 1 vφ = √ , µε (25) where µ = µr µ0 and ε = εr ε0. The refractive index n is defined by: c √ n= = µr εr . vφ (26) Generally, vφ < c, so n > 1. At certain frequencies, resonance can occur between the electric field in the wave, and the oscillation of electrons in the atoms and molecules of the dielectric. In such regimes, the refractive index can become a complex function of the frequency. Electromagnetism 10 Lecture 2: Electromagnetic Waves Electromagnetic Waves in Conductors Electromagnetic waves in conductors do behave differently in some important respects to waves in free space and in dielectrics. The differences arise from the fact that there will be electric currents in the conductor driven by the electric field. For an ohmic conductor, the current density is proportional to the electric field: ~ J~ = σ E. (27) Maxwell’s equation (4) then becomes: ~ ∂E ~ ~ ∇ × B = µσ E + µε . (28) ∂t Proceeding as before, we take the curl, and then substitute for ~ from (3), to find: ∇×E ~ ~ ∂B ∂ 2B 2 ~ ∇ B − µσ − µε 2 = 0. ∂t ∂t Electromagnetism 11 (29) Lecture 2: Electromagnetic Waves Electromagnetic Waves in Conductors Similarly, starting from (3), we find: ~ − µσ ∇2 E ~ ~ ∂E ∂ 2E − µε 2 = 0. ∂t ∂t (30) Note the presence of the first derivative with respect to time. This represents a “damping” term, that will lead to decay of the wave amplitude. The solutions to the wave equations (29) and (30) can be written: ~ r i(~ ~ = B ~ 0e−β·~ B e α·~r−ωt), ~ r i(~ ~ = E ~ 0e−β·~ E e α·~r−ωt). (31) (32) Substituting this solution into the wave equation gives the dispersion relations: ~ = µσω, 2~ α·β (33) ~ 2 = µεω 2. α ~2 + β 12 Electromagnetism (34) Lecture 2: Electromagnetic Waves Electromagnetic Waves in Conductors The dispersion relations are more complicated in a conductor than in free space. In a conductor, the phase velocity is a function of the frequency: s 1 2 |~ α| 1 √ 1 = µε + 1+ 2 2 . ω 2 2 ω ε σ2 (35) We can define a “good conductor” as one for which: σ ωε. (36) (Note that this condition depends on the frequency of the wave). Then, we can make the approximation for the phase velocity: r µσ |~ α| ≈ . (37) ω 2ω Electromagnetism 13 Lecture 2: Electromagnetic Waves Electromagnetic Waves in Conductors: Skin Depth The amplitude of the wave in a conductor falls by a factor 1/e ~ in a distance 1/ β : this defines the skin depth, δ. The skin depth can be found from the dispersion relations (33) and (34). In a good conductor (σ ωε), the skin depth is given approximately by: 1 δ = ≈ ~ β s 2 . µσω (38) Note that this distance is shorter than the wavelength by a factor 2π: s 2π 2 λ= ≈ 2π . |~ α| µσω 14 Electromagnetism (39) Lecture 2: Electromagnetic Waves Electromagnetic Waves in Conductors As well as a decay in the amplitude of the wave in a conductor, there is also a phase difference between the electric and magnetic fields. This follows from the relationship between the amplitudes, which can be found from Maxwell’s equation (3): ~ ~ = −∂B . ∇×E ∂t With the solutions (31) and (32), we find: ~ ×E ~ 0 = iω B ~ 0. i~ α−β (40) For a good conductor (σ ωε), we have: r µσω ~ |~ α| ≈ β ≈ . 2 Hence, the ratio of field amplitudes is given by: Electromagnetism s ~ E0 ω ≈ (i − 1) . ~ 2µσ B 0 15 (41) (42) Lecture 2: Electromagnetic Waves Electromagnetic Waves in Conductors In a good conductor: • the wave amplitude decays exponentially over a distance shorter than the wavelength; • the electric and magnetic fields are out of phase by approximately 45◦. Electromagnetism 16 Lecture 2: Electromagnetic Waves Boundary Conditions on Electromagnetic Fields The electromagnetic waves we encounter in accelerators generally exist within bounded regions, for example within vacuum chambers, rf cavities, or waveguides. The presence of boundaries imposes conditions on the waves that can exist within a given region: often, only particular frequencies and wavelengths are allowed for waves in a bounded region. This is in contrast to waves in free space, where any frequency of wave is allowed. To understand the constraints on waves in bounded regions, we first have to understand the constraints on the fields themselves at boundaries. These “boundary conditions” can be derived from Maxwell’s equations. Electromagnetism 17 Lecture 2: Electromagnetic Waves ~ Boundary Conditions 1: Normal Component of B ~ at First, consider the magnetic field B the boundary between two media. The field must satisfy Maxwell’s equation, in integral form: Z ∂V ~ · dS ~ = 0. B (43) If we take the region V with boundary ∂V to be a thin “pill box” just crossing the boundary, then, taking the limit that the width of the pill box approaches zero, we find the boundary condition: B2⊥ − B1⊥ = 0, (44) where B2⊥ is the normal component of the magnetic field on one side of the boundary, and B1⊥ is the normal component on the other side of the boundary. 18 Electromagnetism Lecture 2: Electromagnetic Waves ~ Boundary Conditions 2: Parallel Component of E ~ at Next, consider the electric field E a boundary. The field must satisfy Maxwell’s equation, in integral form: ∂ ~ · d~ ~ · dS. ~ E `=− B (45) ∂t S ∂S If we take the area S with boundary ∂S to be a thin strip just crossing the boundary, then, taking the limit that the width of the strip approaches zero, we find the boundary condition: Z Z E2k − E1k = 0, (46) where E2k is the parallel component of the electric field on one side of the boundary, and E1k is the parallel component on the other side of the boundary. Electromagnetism 19 Lecture 2: Electromagnetic Waves ~ Boundary Conditions 3: Normal Component of D Now, consider the electric displace~ at a boundary. The field must ment D satisfy Maxwell’s equation, in integral form: Z ∂V ~ · dS ~ = q, D (47) where q is the electric charge enclosed by the surface ∂V . If we take the region V with boundary ∂V to be a thin “pill box” just crossing the boundary, then, taking the limit that the width of the pill box approaches zero, we find the boundary condition: D2⊥ − D1⊥ = ρs, (48) where ρs is the surface charge density (i.e. the charge per unit area) existing on the boundary between the media. 20 Electromagnetism Lecture 2: Electromagnetic Waves ~ Boundary Conditions 4: Parallel Component of H Finally, consider the magnetic inten~ at a boundary. The field must sity H satisfy Maxwell’s equation, in integral form: ~ ∂D ~ · d~ H `= J~ + ∂t ∂S S Z Z ! ~ · dS. (49) If we take the area S with boundary ∂S to be a thin strip just crossing the boundary, then, taking the limit that the width of the strip approaches zero, we find the boundary condition: H2k − H1k = Js, (50) where Js is the surface current density flowing on the boundary, perpendicular to the magnetic intensity. If the media have finite conductivity, we can expect this surface current density to be zero. Electromagnetism 21 Lecture 2: Electromagnetic Waves Laws of Reflection and Refraction The boundary conditions applied to the fields in electromagnetic waves lead to the familiar phenomena associated with reflection and refraction, including: • the law of reflection, and the law of refraction (Snell’s law); • the intensity of reflected and transmitted waves, as functions of the angle of incidence and polarisation of the wave (Fresnel’s equations); • total internal reflection (critical angle); • polarisation by reflection (Brewster angle). However, we will be mostly concerned with waves on the surface of a “good” conductor... 22 Electromagnetism Lecture 2: Electromagnetic Waves Boundary Conditions on the Surface of a Good Conductor Electromagnetic waves at the surface of good conductors (including superconductors) require rather careful analysis. However, the significant phenomena (for our purposes) can be understood if we assume that the skin depth approaches zero as the conductivity becomes infinite. In this approximation, electromagnetic waves cannot penetrate into a good conductor, and we can assume that the fields inside the conductor are zero. Then, applying the boundary conditions, we find (for medium 1 air or vacuum, and medium 2 a good conductor): B1⊥ E1k D1⊥ H1k = = = = 0 0 −ρs Js B2⊥ E2k D2⊥ H2k = = = = 0 0 0 0 Physically, the charges at the boundary respond to the fields in any incident wave to suppress any fields within the conductor. Electromagnetism 23 Lecture 2: Electromagnetic Waves Electromagnetic Waves in a Metal Box Consider a vacuum within a rectangular box with side lengths a, b and d, and perfectly conducting walls. We know that within the cavity, the electromagnetic fields must satisfy the wave equations: ~ − µ0ε0 ∇2 E ~ ∂ 2E = 0, ∂t2 (51) ~ ∂ 2B = 0. (52) ∂t2 However, now we also have to satisfy the boundary conditions on the walls of the box. ~ − µ0ε0 ∇2 B Electromagnetism 24 Lecture 2: Electromagnetic Waves Electromagnetic Waves in a Metal Box The appropriate solutions for the electric field take the form of standing waves: Ex = Ex0 cos kxx sin ky y sin kz z e−iωt Ey = Ey0 sin kxx cos ky y sin kz z e−iωt (53) Ez = Ez0 sin kxx sin ky y cos kz z e−iωt (55) (54) where the wave equation is satisfied if: kx2 + ky2 + kz2 = µ0ε0ω 2. ~ = 0 is satisfied if: Maxwell’s equation ∇ · E kxEx0 + ky Ey0 + kz Ez0 = 0. (56) (57) And finally, the boundary condition (that the parallel component of the electric field vanishes at the walls of the box) is satisfied if: π π π kx = ` , k y = m , k z = n , (58) a b d for integer `, m and n. Electromagnetism 25 Lecture 2: Electromagnetic Waves Electromagnetic Waves in a Metal Box Equations (56), (57), and (58) have important consequences. First, equations (56) and (58) mean that the (standing) waves within the box can only take particular, resonant frequencies: ω`mn = πc √ where c = 1/ µ0ε0. s `2 m2 n2 + + , a2 b2 d2 (59) From equation (57) we see that at least two of `, m and n must be non-zero, otherwise the wave vanishes entirely. Thus, the lowest frequency wave that can exist within the box has frequency (assuming a < b and a < d): ω011 = πc s 26 Electromagnetism 1 1 + . b2 d2 (60) Lecture 2: Electromagnetic Waves Electromagnetic Waves in a Metal Box The (0, 1, 1) mode has electric field given by: Ex = Ex0 sin ky y sin kz z e−iωt, Ey = 0, Ez = 0, (61) (62) (63) where: π π , and kz = . (64) b d The magnetic field is related to the electric field by Maxwell’s equations. (Formally, we have to verify that our solutions satisfy all of Maxwell’s equations and the boundary conditions simultaneously.) For the (0, 1, 1) mode, the magnetic field is given by: ky = Bx = 0, (65) i kz Ex0 sin ky y cos kz z e−iωt, ωµ i Bz = ky Ex0 cos ky y sin kz z e−iωt. ωµ By = − Electromagnetism 27 (66) (67) Lecture 2: Electromagnetic Waves Electromagnetic Waves in a Metal Box Each electromagnetic field “pattern” (with its own frequency) is known as a “mode”. Electromagnetic field modes within a metal cavity are analogous to standing waves or modes that can exist on a stretched wire fixed at each end. Mode (0, 1, 1) Electromagnetism Mode (1, 2, 1) 28 Lecture 2: Electromagnetic Waves RF Cavities in Accelerators RF cavities are commonly used in accelerators to provide an accelerating (longitudinal electric) field. Issues include: • geometry design to maximize the field on-axis, while minimizing the field near the walls (which can lead to ohmic losses); • design of the “couplers” that carry electromagnetic energy into the cavity to excite the desired mode; • design of “higher order mode dampers” that extract electromagnetic energy from undesired modes. Electromagnetism 29 Lecture 2: Electromagnetic Waves Waves in Waveguides In its simplest form, a waveguide is simply a long metal pipe. However, in accelerators, waveguides are extremely useful for carrying electromagnetic energy in the form of waves from a source (e.g. a klystron) to an rf cavity. 30 Electromagnetism Lecture 2: Electromagnetic Waves Waves in Waveguides The analysis of the fields in a rectangular waveguide is similar to the analysis of fields in a rectangular cavity... except that one dimension is extended to infinity. In the “open” dimension, the solution to the wave equation takes the form of a travelling wave; in the bounded directions, the solution has the form of a standing wave. Ex = Ex0 cos kxx sin ky y ei(kz z−ωt), (68) Ey = Ey0 sin kxx cos ky y ei(kz z−ωt), (69) Ez = iEz0 sin kxx sin ky y ei(kz z−ωt), (70) where: π kx = ` , a Electromagnetism and 31 π ky = m . b (71) Lecture 2: Electromagnetic Waves Waves in Waveguides Note that the boundary conditions impose constraints on kx and ky . In principle, there is no constraint on kz , and for given (integer) mode numbers ` and m, any frequency ω is allowed. However, the wave equation leads to the relationship: ω2 2 kz = 2 − kx2 − ky2. c (72) If ω is too small, then kz2 will be less than zero; kz will be an imaginary number. In this case, the wave does not propagate along the waveguide, but instead the fields decay exponentially with distance along z. The lowest frequency for a wave to propagate along the waveguide is known as the cut-off frequency, and is given (for b < a) by: π ωcut-off = c. (73) a 32 Electromagnetism Lecture 2: Electromagnetic Waves Phase Velocity of Waves in Waveguides The phase velocity of a wave in a waveguide is given by: ω = vp = kz q kx2 + ky2 + kz2 kz c. (74) It appears that the phase velocity is greater than the speed of light in vacuum. This is perfectly legitimate, since energy propagates at the group velocity rather than the phase velocity... and as we shall see, the group velocity is less than the speed of light. Electromagnetism 33 Lecture 2: Electromagnetic Waves Phase Velocity and Group Velocity Consider a wave with frequency ω and wave number k: y = cos(kz − ωt). (75) Suppose we superpose another wave on top of this one; but the new wave has slightly different frequency, ω + dω and wave number k + dk (dω ω, and dk k). The total “displacement” is given by: y = cos(kz − ωt) + cos ((k + dk)z − (ω + dω)t) dω dk )t cos(dk z − dω t). = 2 cos (k + )z − (ω + 2 2 (76) We see that the wave is now modulated; the velocity of the modulation is: dω vg = . (77) dk This is the group velocity. Since the energy in a wave is proportional to the square of the amplitude, the energy in a wave propagates at the group velocity. Electromagnetism 34 Lecture 2: Electromagnetic Waves Group Velocity of Waves in Waveguides Zero dispersion: ω = vpk Electromagnetism 35 Lecture 2: Electromagnetic Waves Group Velocity of Waves in Waveguides q With dispersion: ω = vp k2 + α2 36 Electromagnetism Lecture 2: Electromagnetic Waves Group Velocity of Waves in Waveguides The dispersion relation for a wave in a waveguide is, from (72): q ω = c kx2 + ky2 + kz2. Hence, the group velocity is: (78) ckz ∂ω =q . (79) ∂kz kx2 + ky2 + kz2 For real values of kx and ky , the group velocity is less than the speed of light. In fact, we find that vpvg = c. vg = For given mode numbers ` and m, we can plot the group velocity as a function of frequency: Electromagnetism 37 Lecture 2: Electromagnetic Waves Impedance We have seen that it is possible to find analytical solutions to Maxwell’s equations in bounded regions with simple geometries. In more realistic situations, where the geometries involved are more complicated, it is usually necessary to solve Maxwell’s equations numerically, using a specialised computer code. However, one further important situation that is amenable to analysis concerns waves incident on boundaries. We will not treat the general case, but only consider normal incidence. This introduces the important concept of impedance. Electromagnetism 38 Lecture 2: Electromagnetic Waves Electromagnetic Waves Normally Incident on a Boundary For simplicity, we shall consider an infinite plane electromagnetic wave striking a boundary between two dielectrics at normal incidence. To satisfy the boundary conditions, we need, as well as the incident wave, a transmitted and a reflected wave. Electromagnetism 39 Lecture 2: Electromagnetic Waves Electromagnetic Waves Normally Incident on a Boundary Write the electric fields and the magnetic intensities in the waves as: HI = EI0 i(kz−ωt) e Z1 ER = ER0ei(kz+ωt) HR = ER0 i(kz+ωt) e Z1 ei(kz−ωt) HT = ET 0 i(kz−ωt) e Z2 EI = EI0 ei(kz−ωt) ET = ET 0 Here, we have used the fact that the amplitudes of the electric and magnetic fields in plane wave are related by: E0 = H0 r µ = Z. ε (80) Z is the impedance of the medium through which the wave is travelling. 40 Electromagnetism Lecture 2: Electromagnetic Waves Electromagnetic Waves Normally Incident on a Boundary Now we use the boundary conditions, which tell us that the ~ and H ~ must be continuous across tangential components of E the boundary. These conditions can be written: ET = EI + ER , and HT = HI − HR , (81) where the minus sign in the equation for the magnetic intensity takes into account the direction of the field in the reflected wave. We find that the boundary conditions are satisfied if: 2Z2 ET 0 = , EI0 Z1 + Z2 and Z − Z1 ER0 = 2 . EI0 Z2 + Z1 (82) Hence, the amplitudes of the reflected and transmitted waves are determined by the impedances of the media on either side of the boundary. Note that if Z1 = Z2, then the wave propagates as if there was no boundary. Electromagnetism 41 Lecture 2: Electromagnetic Waves Impedance Although we have dealt with a very special case, the concept of impedance is used very widely in electromagnetism. In general, impedance is used to relate: • the ratio of the electric field to the magnetic intensity in an electromagnetic wave; • the ratio of voltage to current in a circuit; • the voltage acting back on a beam in an accelerator as a result of the beam current. Any discontinuity in a medium or circuit through which energy is flowing as a wave will be associated with a change in impedance. Knowing the impedance is important for understanding the transmission and reflection of energy at the discontinuity. Electromagnetism 42 Lecture 2: Electromagnetic Waves Sources of Electromagnetic Radiation If a point charge, initially at rest, is moved from one place to another, changes in the field around that charge will propagate through space at the speed of light. The “disturbance” in the electromagnetic field carries energy. Electromagnetic radiation is emitted whenever electric charges undergo acceleration. Electromagnetism 43 Lecture 2: Electromagnetic Waves Sources of Electromagnetic Radiation The Hertzian dipole is one of the simplest “radiating systems” to analyse. This consists of an infinitesimal pair of point charges, oscillating so that the electric dipole moment is: ~ =p p ~ = q ∆z ~0e−iωt. (83) The oscillation may be represented as an alternating current flowing between two fixed points. The vector potential can then be found from: ~ r, t) = µ0 A(~ 4π Z ~" 0 J ~ r , t − |~ r−~ r 0|/c dV 0. 0 |~ r−~ r | (84) The magnetic field is found from (5): ~ = ∇ × A, ~ B and the electric field from Maxwell’s equation (3): ~ ~ = −∂B . ∇×E ∂t Electromagnetism 44 Lecture 2: Electromagnetic Waves Sources of Electromagnetic Radiation We do not go into the detailed analysis of the Hertzian dipole. However, the result is that the dipole radiates energy, that (in the mid-plane and at large distances from the dipole) looks like plane waves. For more details, see e.g. Grant and Phillips. Understanding sources of radiation is important in accelerators, particularly in the context of synchrotron radiation. Electromagnetism 45 Lecture 2: Electromagnetic Waves Summary You should be able to: • derive, from Maxwell’s equations, the wave equations for the electric and magnetic fields in free space, dielectrics, conductors; • derive the principal properties of the waves, including: speed of propagation, relative directions and amplitudes of the fields, and (in conductors) the skin depth; • derive the solutions to the wave equations in regions bounded by conductors; • explain the properties of electromagnetic waves in rf cavities and waveguides, and discuss how these components are useful in accelerators; • explain that electromagnetic waves are generated whenever charges undergo acceleration. Electromagnetism 46 Lecture 2: Electromagnetic Waves